重写路径段的任意数量的查询参数 [英] Rewriting an arbitrary number of path segments to query parameters

问题描述

我有这样的.htaccess规则：

 重写规则viewshoplatest /(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/ （。*）/（。*）/（。*）/（。*）/（。*）/（。*）/（。*）/（。*）/（。*）/ $ /viewshoplatest.php?$1=$2&$3=$4&$5=$6&$7=$8&$9=$10&$11=$12&$13=$14&$15=$16
 

它应该映射的URL是这样的：

<$p$p><$c$c>http://www.veepiz.com/viewshoplatest/start/10/end/10/filter/0/ownerid/0/sortby/date/sortdir/DESC/cat/0/scat/0/

这样：

<$p$p><$c$c>http://www.veepiz.com/viewshoplatest.php?start=0&end=10&filter=0&ownerid=0&sortby=date&sortdir=DESC&cat=0&scat=0

在我的链接，点击打印 $ _ GET 变量，我得到这样的：

 阵列（[开始] =＆GT; 10月底] =＆GT; 10 [过滤器] =＆GT; 0 OWNERID] =＆GT; 0 sortby] =＆GT; START0 [启动1] =＆GT; START2 [start3] =＆GT; start4 [start5] =＆GT; start6）
 

任何想法，为什么它的表现不好？

---

好吧，我已经通过重写规则来解决了这个

 重写规则viewshoplatest/start/(.*)/end/(.*)/filter/(.*)/ownerid/(.*)/sortby/(.*)/sortdir/(.*)/cat/(.*)/scat/(.*)/$ /viewshoplatest.php?start=$1&end=$2&filter=$3&ownerid=$4&sortby=$5&sortdir=$6&cat=$7&scat=$8
 

解决方案

首先：你不应该使用 * 如果你能更具体，非常相似。这种情况下， [^ /] + 。由于多个。* 可造成巨大的回溯。所以：

 重写规则^viewshoplatest/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/$ /viewshoplatest.php?$1=$2&$3=$4&$5=$6&$7=$8&$9=$10&$11=$12&$13=$14&$15=$16
 

您可以使用一个像花使用RegexBuddy 看到这些常规的前pressions如何处理的差异。

但是，因为mod_rewrite的确实只允许引用第一九组（见<一href=\"http://stackoverflow.com/questions/3655893/another-mod-rewrite-problem/3656017#3656017\">Tim’s回答），你可以使用一个迭代的方法，并可以同时处理一个参数：

 重写规则^ viewshoplatest /([^/]+)/([^/]+)/([^/]+/[^/]+/.*)$ / viewshoplatest / $ 3〜$ 1 = $ 2 [QSA，N]
重写规则^ viewshoplatest /（[^ /] +）/（[^ /] +）/（[^ /] *）/ $ /viewshoplatest.php?$1=$2&$3 [QSA，L]
 

第一条规则将处理在同一时间（除了最后一对）一个参数对通过追加到已经存在的（见的 QSA 的标志），然后重新开始重写操作不增加内部递归计数器（见的 N 的标志）。然后第二个规则将改写最后一个参数对（或只是名称），结束迭代。

但由于使用的 N 的标志可能是危险的，因为它会导致无限递归，你也可以使用PHP来解析请求的路径：

  $ _ SERVER ['REQUEST_URI_PATH'] = parse_url（$ _ SERVER ['REQUEST_URI']，PHP_URL_PATH）;
$段=破灭（'/'，修剪（$ _ SERVER ['REQUEST_URI_PATH']，'/'））;
array_shift（$段）; //删除路径preFIX/ viewshoplatest
为（$ i = 0，$ N =计数（$段）; $ I＆LT; $ N;）{
    $ _GET [rawurlde code（$段[$ I ++]）=（$ I＆LT; $ N）？ rawurlde code（$段[$ I ++]）：空;
}
 

现在你只需要这条规则通过将请求传递：

 重写规则^ viewshoplatest（/ | $）/viewshoplatest.php [L]
 

I have this .htaccess rule:

RewriteRule viewshoplatest/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/(.*)/$ /viewshoplatest.php?$1=$2&$3=$4&$5=$6&$7=$8&$9=$10&$11=$12&$13=$14&$15=$16

It should map a URL like this:

http://www.veepiz.com/viewshoplatest/start/10/end/10/filter/0/ownerid/0/sortby/date/sortdir/DESC/cat/0/scat/0/

to this:

http://www.veepiz.com/viewshoplatest.php?start=0&end=10&filter=0&ownerid=0&sortby=date&sortdir=DESC&cat=0&scat=0

When I click on link and print $_GET variables I get this:

Array ( [start] => 10 [end] => 10 [filter] => 0 [ownerid] => 0 [sortby] => start0 [start1] => start2 [start3] => start4 [start5] => start6 )

Any ideas as to why it’s behaving badly?

---

Ok i have fixed this by rewriting the rule to

RewriteRule viewshoplatest/start/(.*)/end/(.*)/filter/(.*)/ownerid/(.*)/sortby/(.*)/sortdir/(.*)/cat/(.*)/scat/(.*)/$ /viewshoplatest.php?start=$1&end=$2&filter=$3&ownerid=$4&sortby=$5&sortdir=$6&cat=$7&scat=$8

解决方案

First of all: You shouldn’t use .* if you can be more specific, like in this case [^/]+. Because multiple .* can cause immense backtracking. So:

RewriteRule ^viewshoplatest/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/([^/]+)/$ /viewshoplatest.php?$1=$2&$3=$4&$5=$6&$7=$8&$9=$10&$11=$12&$13=$14&$15=$16

You can use a took like RegexBuddy to see the difference in how these regular expressions are processed.

But since mod_rewrite does only allow to reference the first nine groups (see Tim’s answer), you could use an iterative approach and process one parameter at a time:

RewriteRule ^viewshoplatest/([^/]+)/([^/]+)/([^/]+/[^/]+/.*)$ /viewshoplatest/$3?$1=$2 [QSA,N]
RewriteRule ^viewshoplatest/([^/]+)/([^/]+)/([^/]*)/?$ /viewshoplatest.php?$1=$2&$3 [QSA,L]

The first rule will process one parameter pair at a time (except the last pair) by append it to the already existing ones (see QSA flag) and then restart the rewriting process without incrementing the internal recursion counter (see N flag). The second rule will then rewrite the last parameter pair (or just the name) and end the iteration.

But since using the N flag might be dangerous as it can cause an infinite recursion, you could also use PHP to parse the requested path:

$_SERVER['REQUEST_URI_PATH'] = parse_url($_SERVER['REQUEST_URI'], PHP_URL_PATH);
$segments = implode('/', trim($_SERVER['REQUEST_URI_PATH'], '/'));
array_shift($segments); // remove path prefix "/viewshoplatest"
for ($i=0, $n=count($segments); $i<$n; ) {
    $_GET[rawurldecode($segments[$i++])] = ($i < $n) ? rawurldecode($segments[$i++]) : null;
}

Now you just need this rule to pass the request through:

RewriteRule ^viewshoplatest(/|$) /viewshoplatest.php [L]

这篇关于重写路径段的任意数量的查询参数的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！